Project Summary Understanding the molecular and cellular basis for behavior depends on a rigorous assessment of the contributions of different neuron types. Progress in elucidating neural circuits for behaviors is often hampered by a lack of genetic tools for efficiently generating animals with cell-type specific expression of genes that can be used to perturb or monitor neuronal activity, such as optogenetic tools, tetanus toxin, and GCaMP. Also, there has been steady progress in optogenetics and genetically-encoded sensors such as GCaMP calcium indicators, but it is impractical to rebuild hundreds of strains inserting each improved version into a repertoire of expression vectors. One elegant method that addresses both issues simultaneously is to use a bipartite expression system, which separates the cell-type control from the effector, and thus a set of cell-type specific drivers can be reused with different versions of effectors (e.g., GCaMP6 versus GCaMP3). Conversely, a set of strains might be made with a promoter that directs expression in two or more cell types; if a more specific regulatory sequence is identified, then all the constructs have to be rebuilt. With a bipartite system, construction of a single Driver with the new regulatory sequence can easily combined with all the available Effectors to efficiently generate the strains needed. allows use of all the Effectors. For example, Drosophila researchers have made great use of the Gal4-UAS system in which a transcriptional activator protein (Gal4) is expressed in the cell type(s) of interest and binds to its target sequence ? the UAS ? to direct expression of an effector gene of interest. This scheme allows many combinations of specifically expressed genes to be built from a much smaller number of transgenes. However, Caenorhabditis elegans has not had such a system until our recent development of the cGAL system, an optimized Gal4-UAS system. One key feature of our implementation is the use of the DNA-binding domain of the Gal4 protein from a yeast species whose optimal growth temperature matches that of C. elegans, thereby allowing more efficient target gene activation. We also showed that the cGAL system can be applied to functional studies in C. elegans. We propose to construct an initial neuronal cGAL toolkit, and apply it to one circuit as proof of principle. The chosen circuit is male mating behavior, arguably the most complex of C. elegans behaviors as it involves almost the entire nervous system and a complex series of steps each involving sensory-motor integration. While the roles of many male specific neurons have been identified, the roles of non-sex-specific neurons have not; our approach will make the cGAL reagents that render all of the non-sex-specific neurons tractable to analysis in a systematic way. At the end of two years, we will have fully introduced a useful bipartite expression system to the C. elegans community and refined our understanding of innate behavior.